Triadic Coercion: Israel’s Targeting of States That Host Nonstate Actors 9780231548540

Citation preview

T R I A D I C COE R CI O N

Columbia Studies in Terrorism and Irregular Warfare

COLUMBIA STUDIES IN TERRORISM AND IRREGULAR WARFARE Bruce Hoffman, Series Editor This series seeks to fill a conspicuous gap in the burgeoning literature on terrorism, guerrilla warfare, and insurgency. The series adheres to the highest standards of scholarship and discourse and publishes books that elucidate the strategy, operations, means, motivations, and effects posed by terrorist, guerrilla, and insurgent organizations and movements. It thereby provides a solid and increasingly expanding foundation of knowledge on these subjects for students, established scholars, and informed reading audiences alike. Ami Pedahzur, The Israeli Secret Services and the Struggle Against Terrorism Ami Pedahzur and Arie Perliger, Jewish Terrorism in Israel Lorenzo Vidino, The New Muslim Brotherhood in the West Erica Chenoweth and Maria J. Stephan, Why Civil Resistance Works: The Strategic Logic of Nonviolent Conflict William C. Banks, editor, New Battlefields/Old Laws: Critical Debates on Asymmetric Warfare Blake W. Mobley, Terrorism and Counterintelligence: How Terrorist Groups Elude Detection Jennifer Morrison Taw, Mission Revolution: The U.S. Military and Stability Operations Guido W. Steinberg, German Jihad: On the Internationalization of Islamist Terrorism Michael W. S. Ryan, Decoding Al-Qaeda’s Strategy: The Deep Battle Against America David H. Ucko and Robert Egnell, Counterinsurgency in Crisis: Britain and the Challenges of Modern Warfare Bruce Hoffman and Fernando Reinares, editors, The Evolution of the Global Terrorist Threat: From 9/11 to Osama bin Laden’s Death Boaz Ganor, Global Alert: The Rationality of Modern Islamist Terrorism and the Challenge to the Liberal Democratic World M. L. R. Smith and David Martin Jones, The Political Impossibility of Modern Counterinsurgency: Strategic Problems, Puzzles, and Paradoxes Elizabeth Grimm Arsenault, How the Gloves Came Off: Lawyers, Policy Makers, and Norms in the Debate on Torture Assaf Moghadam, Nexus of Global Jihad: Understanding Cooperation Among Terrorist Actors Bruce Hoffman, Inside Terrorism, 3rd edition Stephen Tankel, With Us and Against Us: How America’s Partners Help and Hinder the War on Terror

Triadic Coercion I S R A E L’ S TA R G E T I N G O F S TAT E S T H AT H O S T N O N S TAT E AC TO R S

Wendy Pearlman and Boaz Atzili

Columbia University Press New York

Columbia University Press Publishers Since 1893 New York Chichester, West Sussex cup.columbia.edu Copyright © 2018 Columbia University Press All rights reserved Library of Congress Cataloging-in-Publication Data Names: Pearlman, Wendy, author. | Atzili, Boaz, author. Title: Triadic coercion : Israel, Arab states, and non-state actors in conflict / Wendy Pearlman and Boaz Atzili. Other titles: Columbia studies in terrorism and irregular warfare. Description: New York : Columbia University Press, 2018. | Series: Columbia studies in terrorism and irregular warfare | Includes bibliographical references and index. Identifiers: LCCN 2018007692 | ISBN 9780231171847 (cloth : alk. paper) | ISBN 9780231548540 (e-book) Subjects: LCSH: Non-state actors (International relations) | International relations. | Security, International. | Arab-Israeli conflict. | Israel—Foreign relations—Arab countries. | Arab countries—Foreign relations—Israel. Classification: LCC JZ4059 .P43 2018 | DDC 956.05/4—dc23 LC record available at https://lccn.loc.gov/2018007692

Columbia University Press books are printed on permanent and durable acid-free paper. Printed in the United States of America Cover design: Lisa Hamm

We dedicate this book to our fathers and also to the struggle for peace and freedom in the Middle East.

Contents

Map of Israel and the Surrounding Region ix Preface and Acknowledgments xi 1 2

3 4

Understanding Triadic Coercion 1

Israel’s Use of Triadic Coercion: Sources and Historical Evolution 23

Egypt Since 1949: Triadic Coercion from Raids to Peace 59 Syria Since 1949: Triadic Coercion from Coups to Revolution 95 5

Israel and the Palestinian Authority Since 1993: Strategic Culture in Asymmetric Conflict 130 6

Lebanon Before and Since 2006: Strategic Culture at War 175 7

Triadic Coercion Beyond the Arab-Israeli Conflict 218 Conclusion

242

Notes 255 Bibliography 313 Index 351 [ vii ]

Map of Israel and the Surrounding Region

Preface and Acknowledgments

T

his project’s roots were planted at Harvard’s John  F. Kennedy School of Government’s Belfer Center for Science and International Affairs, where we had postdoctoral fellowships in 2007– 2008. Having offices across the hallway from each other, we had many conversations about the still-fresh 2006 War between Israel and Hezbollah. We shared a deep dismay about the vast destruction and loss of life, and were particularly alarmed by the scope of Israel’s bombardment of Lebanon. What began as a normative concern about excessive use of force spurred our attention to an intellectual puzzle about its targeting. Why would Israel pound Lebanon, a famously weak state, to demand that it block the actions of Hezbollah, a famously strong nonstate actor? To answer these questions, we delved into both theory and history. Building on work on deterrence, intrastate conflict, and other dynamics of warfare, we crafted an analytical framework and developed generalizable hypotheses about the causes and effects of states’ use or threat of force  to coerce other states to stop attacks by nonstate actors. We then investigated these dynamics across the rich seventy-year trajectory of the Arab-Israeli conflict. Researching the foundations of Israel’s security decision-making and thinking from statehood to the present, we discovered that its approach to fighting nonstate actors was not uniform. Israel had not always targeted host states in the demand that they “take responsibility” for stopping nonstate actors on their soil. Even when it employed [ xi ]

this policy, it did not do so in precisely the same way over time or space. Seeking to explain these shifts on the part of Israel in these three-actor conflicts, we were struck by how developments in its strategic culture went far in accounting for variation and, in particular, for what appeared to be its increasingly aggressive actions in recent years. We also threw ourselves into the study of Arab states that Israel charged with harboring violent nonstate actors. We examined their governing regimes, with particular interest in regime origins, internal compositions, institutional structures, and connections to society. We were attentive to the complex relationships between host regimes and nonstate actors, noting how they could range from patronage to hostility. We saw that these relationships were typically driven less by regimes’ claimed ideological commitments to armed struggle against Israel than they were by concern for their own political survival. In this context, we were drawn to a consistent pattern: only regimes that were strong, in terms of both cohesion and capacity, had the ability to determine when raison d’état demanded that they assert control over nonstate actors and then do what was needed to impose that control. This pattern had direct implications for Israel’s strategy of using military pressure against host states to force them to act against nonstate actors. Such a strategy, history showed us, was bound to fail when that host’s regime was weak. The product of this multistranded research is this book on what we came to call “triadic coercion.” We set out to study this topic with as much scholarly objectivity as possible in the belief that social scientific analysis of the empirical record was the best way to make inferences about causes and effects and thereby derive both academic conclusions and policy recommendations. At the same time, we have not lost sight of the moral intensity of the competing commitments that imbue every aspect of the Arab-Israeli conflict. In beginning our research with the wake of the 1948 War, we do so with attentiveness to the momentousness of that war itself as an historic justice or injustice, depending on one’s point of view. We acknowledge that even the most seemingly simple terms that we employ in this book might be deeply contested. We understand that the violence that we refer to with such dry terms as “attacks,” “reprisals,” “retaliation raids,” and “operations” are not simply events to be modeled and counted but actions that can obliterate lives, destroy families, terrorize or dehumanize entire communities, and rob individuals of their most basic rights to life and physical security. Similarly, the “nonstate actors” in our story are [ xii ]

P R E F A C E A N D A C K N OW L E D G M E N T S

not simply nuisances for state sovereignty and border security. For some they are terrorist groups deliberately killing civilians. For others they are the legitimate representatives of dispossessed peoples struggling for liberation. For the sake of consistency, we refer to the main actors that we examine by the names that they call themselves. When we use Israel’s chosen name for its army, the Israel Defense Forces, we do so with sensitivity to those who view it as a force of aggression and occupation rather than defense. As this book was many years in the making, the list of people to whom we owe gratitude is similarly long. We are thankful to the Belfer Center for the intellectual environment it provided us as newly minted PhDs. We could not have predicted that our casual chats would spawn an article and then an academic monograph but are grateful for institutes like Belfer and others that offer academics the time, resources, and community to explore new ideas and follow them to wherever they lead. In the years since then many others helped us make this project better than we could have made it alone. Early in this project several scholars offered feedback on what became our first write-up about what we then called “triadic deterrence,” an article in the journal Security Studies. We thank Jeff Colgan, Erica Chenoweth, Christopher Day, Patrick Thaddeus Jackson, Amir Lupovici, Joseph Olmert, and Janice Stein for valuable comments and suggestions during that stage. We are particularly grateful to Zeev Maoz, who offered us early advice and encouragement and made his datasets available for our use. As we worked on this project over the course of years, we presented parts of it at workshops and talks at Northwestern University, American University, and the University of Chicago, and the annual conferences of the International Studies Association and Association of Israeli Studies. At each event we were lucky to gather feedback from engaged readers and listeners, all of which left important imprints on this work. We are grateful to our institutions, American University and Northwestern University, for the multiple forms of support they gave us, from library services to time to commit to writing. We consider ourselves fortunate to be able to call these universities home. In addition, we express special thanks to American’s SIS Research, especially Shannon Looney and Holly Bennett Christiansen, for funding and expert organizational help in sponsoring a daylong “book incubator” to discuss an early draft of the complete manuscript. We owe gratitude to Dima Adamsky, Ariel Ahram, Jon P R E F A C E A N D A C K N OW L E D G M E N T S

[ xiii ]

Caverley, Keith Darden, Miles Kahler, Austin Long, and Jeremy Pressman for participating in this workshop. Their insights and brainstorming were invaluable in shaping and improving the project. Similarly, we thank two anonymous readers solicited by Columbia University Press for their comments. Throughout the process of producing this book, we benefitted from research assistance from students at both American University and Northwestern University. For that, we owe gratitude to Laura Bosco, Jiajie He, Min Jung Kim, Karen Saunders, Jeffrey Bilik, Alli Divine, Clara Clymer, John Quigley, and Tova Yampolsky. We could not have asked for a better team. We thank the dedicated staff at Columbia University Press, and especially Caelyn Cobb and Miriam Grossman, for shepherding the manuscript from submission to fruition. Lastly, we owe a great debt to our families and friends. Wendy gives special thanks to her partner, Peter Cole, and her grandmother, Margaret Pearlman. Boaz thanks his son, Tomer, daughter, Melanie, and wife, Orit. These and others dear to our hearts not only bore the burden of the countless months in which we were buried in our books and computer screens but also endured occasional lectures on the concept of strategic culture or now-forgotten skirmishes from the 1950s. For us, their patience and good cheer are proof of all that is good in human nature. We dedicate this book to our fathers. Wendy’s father, Michael Pearlman, passed away just a few weeks before we finished this book manuscript. She misses him tremendously but is grateful for what she learned from his love of history, passion for scholarship, tremendous wit, and inspiring humility. He taught her to try to keep things in perspective and be grateful for her immense good fortune in being able to study for a living. As he used to say, “it sure beats a real job.” Boaz’s father, Avraham Atzili, was wounded at the altar of triadic coercion, though he might not know that this was the larger context in which a bullet hit his leg. Beyond this event, Avraham, always the intellectual and the historian, instilled in his son curiosity about the past and a desire to improve the future.

[ xiv ]

P R E F A C E A N D A C K N OW L E D G M E N T S

T R I A D I C COE R CIO N

CH A P T E R 1

Understanding Triadic Coercion

We will make no distinction between the terrorists who committed these acts and those who harbor them. — GE O RGE  W. BUSH, SE PTE MB ER  11, 2001

B

arely had the Twin Towers fallen on September  11 when U.S. president George W. Bush announced that he held accountable not only the al- Qaeda network that flew the planes but also the state in which that organization was based. “The Taliban must act and act immediately,” he warned the governing authorities of Afghanistan. “They will hand over the terrorists or they will share in their fate.”1 Bush’s declarations were reflective of a general trend in the post– Cold War era. Though an increasing number of conflicts involve states and nonstate actors, states often have difficulty fighting such groups due to their small size, secretive structures, lack of visible assets, and sometimes extremist ideologies. Given these circumstances, some analysts conclude that states cannot deter nonstate actors directly and instead should aim to coerce the states that host them. We call such a strategy “triadic coercion,” wherein one state uses military threats and/or punishments against another state to deter it from aiding or abetting attacks by nonstate actors from within its territory or to compel it to stop such violence.2 According to Idean Salehyan, 55 percent of the 291 rebel groups that were known to be fighting against states between 1951 and 1999 used the territory of a neighboring state as a sanctuary or a base for attacks.3 Facing such attacks, governments then and now have responded by using force or threats against the states that willingly or unwillingly played the role of host. The United States has [1]

continually pressured Pakistan to crack down on the Afghan Taliban and Haqqani networks.4 Turkey has engaged Iraq, Syria, and Iran in its struggle against the Kurdistan Workers’ Party.5 India has repeatedly warned Pakistan to end complicity with Kashmiri separatists.6 Sudan attacked Chad to force it to cease support for Darfurian rebels.7 Nicaragua’s Sandinista government pressured Honduras and Costa Rica to stop harboring Contra rebels.8 Triadic coercion, thus, has been entwined with some of the most pressing conflicts in the contemporary international system. Extending our earlier work,9 this book explores triadic coercion through scrutiny of nearly seventy years of Israeli history. For the entirety of its existence, Israel has engaged in strategies of triadic coercion against neighboring Arab states in an effort to end violence and border violations by nonstate actors from those states. Some might argue that this particularly intensive history renders Israel a unique case not suitable for generalizable insight. By contrast, we posit that the very heightened character of Israel’s conflicts with nonstate actors and host states offers a natural laboratory for evaluating triadic coercion as a strategy. Furthermore, it presents a wealth of within-case variation for causal inference. Our longue durée examination of these rises and falls in Israel’s triadic conflicts makes patterns visible that might be difficult to detect elsewhere. Two main questions drive our research. First, we investigate the conditions under which triadic coercion is likely to succeed. Traditional discussions of interstate conflict assume that the greater a state’s power relative to the state that it seeks to coerce, the more likely coercion is to succeed. The coercer state should thus prefer its adversary to be weak. We argue that triadic coercion turns this logic on its head. For the host state to act effectively against a nonstate actor on its soil, it requires internal political cohesion and institutional capacity. The host state’s executive decision-making must be sufficiently consolidated to recognize the national security interest in averting another state’s coercive assaults. It likewise must possess the competencies to design measures against the nonstate actor and the domestic clout to implement such policies even if they are unpopular. Strong regimes possess these means, whereas weak ones do not. As a result, we argue, triadic coercion can only succeed when directed against a host state with at least a minimum level of regime strength. Triadic coercion strategies against weak hosts are unlikely to suppress nonstate actor violence, regardless of the military power that the coercer state projects.

[2]

U N D E R S TA N D I N G T R I A D I C C O E R C I O N

Our second major query builds on the first. If triadic coercion is only effective against strong regimes, why do states frequently employ it against weak ones? We attribute such actions to strategic culture. A state’s system of beliefs, values, and practices can elevate the use of coercive force as an appropriate response— and host states as appropriate targets—independent of the efficacy of those choices. Politicians, public opinion, media, and military leaders might contribute to rhetoric, attitudes, images, and analogies that crystalize this strategic culture. Under these circumstances, national security decision-making elite, and perhaps even the social and political environment at large, encourage triadic coercion on the basis of conviction rather than calculation, and due to a utility that is intrinsic rather than instrumental. The outcome is increased use of the strategy, even when conditions for success are absent. In investigating these questions, we adopt an approach of “analytical eclecticism”: a selective integration of concepts, logics, and mechanisms that stem from separate paradigms as needed to address different aspects of the substantive problem under consideration.10 Our analysis of the sources of triadic coercion policy in Israel’s early years uses a mix of arguments, ranging from a straightforward realist orientation to a more liberal emphasis on domestic politics. Noting that such rationalist lenses fail to explain the persistence over time of policies when they are unlikely to be effective, we find greater explanatory value in an alternative perspective. Our study of the sources of triadic coercion in more recent years thus adopts a constructivist sensitivity to the ways in which collective perceptions, beliefs, and culture are crucial for understanding strategic choices. Employing this varied approach, this introductory chapter elaborates on these arguments, lays out an analytical framework for evaluating them, and discusses the broader contribution of this research.

Triadic Coercion Triadic conflicts occur when a nonstate actor that is geographically based in a “host” state attacks a second state. The attacked state plays what we identify as the role of the “coercer” when it carries out threats or military operations that attempt to persuade the host to prevent the nonstate actor’s attacks (see figure 1.1).

U N D E R S TA N D I N G T R I A D I C C O E R C I O N

[3]

Host State

Coercer State Nonstate Actor

Figure 1.1 Triadic coercion.

Confl icts often become triadic when coercer states conclude that they are unable to resolve their disputes with nonstate actors using bilateral strategies targeting those actors directly. Among such bilateral strategies, a coercer state might attempt to defeat the nonstate actor or render it inoperable by targeting its fighters and assets. Such an approach is often termed “brute force.”11 If unable to defeat the nonstate actor with brute force, the state might seek to use various kinds of deterrence. The state might attempt “deterrence by denial,” through which it seeks to minimize nonstate actors’ attacks by preventing their effective execution. Approximating regular defense strategies, deterrence by denial may involve gathering intelligence, erecting physical barriers, or engaging in diplomatic cooperation to enlist the host state’s help.12 Alternatively, a state might pursue “deterrence by punishment.” The state then uses threats or applies force to demonstrate to the nonstate actor that the costs of aggression outweigh its benefits. Though a rational actor might cease attacks, many analysts doubt that nonstate actors would necessarily do so. Some posit that militant groups are motivated by extremist ideologies that state leaders simply cannot affect. Even if terrorism is judged to be a strategic rather than fanatical choice,13 the rationales of groups that employ it may entail preferences, calculations, and decisionmaking processes that are distinct from those of states. This makes it more difficult for states to communicate threats in a way that will coerce [4]

U N D E R S TA N D I N G T R I A D I C C O E R C I O N

nonstate actors, undermining the prospects that nonstate actors will behave as the logic of deterrence assumes.14 Nonstate actors are also well positioned to withstand deterrence by punishment due to their organizational characteristics. Small size and secretive structures may allow them to elude state intelligence. Nonstate actors typically lack valuable assets to threaten and are unlikely to have a territorial “return address” against which to retaliate.15 Paradoxically, the asymmetry of these confl icts might thus work to the detriment of states. The balance of motivation thus may favor the nonstate actor—whose very survival is at stake—even if the balance of power favors the state.16 Some propose that the nonstate actor might not even have an interest in averting harsh retaliation. According to that logic, nonstate actors might in fact welcome punishment in the belief that it will increase popular sympathy for their cause.17 Given these difficulties surrounding direct deterrence by punishment, a state seeking to defeat a nonstate actor might instead pursue “indirect deterrence.” Such a strategy, Jeff rey Knopf explains, is “not aimed at the attackers themselves but at third parties whose action can affect the likelihood that a potential attacker can or will carry out an assault.”18 Those third parties are typically other states. As nonstate actors do not enjoy their own sovereignty, they must be based within the territory of some state. In addition to territorial safe haven, these groups often seek material assistance, which states are best positioned to supply. Recognizing this, some argue that the principal way for states to fi ght violent nonstate actors is via the states from which they operate or obtain support.19 We focus on “coercer states” that employ this strategy of triadic coercion against “host states” on whose territory nonstate actors are based. We recognize that states can aid insurgent or terrorist groups in ways other than hosting them. However, as Daniel Byman writes, physical haven is “perhaps the most important form of support” that states can provide.20 Unlike other types of sponsorship that entail significant resources, any state can be a host state simply by virtue of possessing territory. Investigation of triadic coercion against host states is thus potentially applicable to a large universe of cases. Triadic coercion against host states entails a mix of two strategic logics. It seeks deterrence by invoking threat of punishment, or limited strikes that suggest the possibility of harsher punishment, to dissuade host states from enabling nonstate actors’ violence. It also entails “compellence” to force U N D E R S TA N D I N G T R I A D I C C O E R C I O N

[5]

the host state to reverse some actions, such as assisting the nonstate actor, or begin taking new actions, such as repressing the nonstate actor or impeding its free movement.21 Conventional wisdom about the difficulty of combatting nonstate actors suggests that a coercer state is rational to target deterrence and compellence at host states instead.22 Keren Fraiman, for example, argues that “placing the onus of containing violent groups back on the states that host them is an attractive and a potentially efficacious strategy.”23 Still, triadic coercion is far from easy. Byman notes: “Support for terrorism can be exceptionally difficult to stop. Sponsors often anticipate the punishment that they may receive for backing terrorists and nevertheless choose to provide support, believing they can endure or avoid the pain. . . . Military strikes— particularly limited ones— often backfire.”24 Jonathan Shimshoni concurs. Scrutinizing some of the same empirical material that we do, he concludes that a state’s attempts to coerce a host state are likely to fail if that host is either sufficiently strong to confront the coercer or too weak or sympathetic to the target group to act against it.25 Shimshoni introduces this observation in the final pages of his study and does not have a chance to develop it into a set of general arguments, no less evaluate them systemically across time and space. Taking up where Shimshoni left off, we craft a study centered on two questions: Under what conditions can triadic coercion succeed? Why do states pursue it even when those conditions do not hold?

Outcomes: Under What Conditions Can Triadic Coercion Succeed? We conceptualize triadic coercion outcomes along a spectrum, assuming the point of view and goals of the coercer state (see figure 1.2). At one extreme, complete success is achieved when the coercer’s actions toward the host state causes it to take actions that result in total cessation of the nonstate actor’s belligerence. At the other extreme, triadic coercion is counterproductive when the coercer’s actions toward the host state cause it to take actions that intensify the nonstate actor’s belligerence. Midway between these poles, triadic coercion can be viewed as irrelevant when the coercer’s attempts to pressure the host state have no effect on the nonstate actor. On the continuum from irrelevance to counterproductivity [6]

U N D E R S TA N D I N G T R I A D I C C O E R C I O N

Figure 1.2 Spectrum of triadic coercion outcomes.

are different levels of strategic failure that result in increasingly greater threats to the coercer. On the continuum from irrelevance to complete success are outcomes of partial success that offer increasingly greater benefits for the coercer. To predict the likely outcomes of triadic coercion, one must understand interstate and intrastate dynamics, which operate at three levels (see figure 1.3). Figure 1.3 presents a decision tree that details the logic of the argument elaborated and evaluated throughout this book. Again, this tree assumes the point of view of the coercer state. The first level is the issue of greatest salience in interstate relations: balance of power. If the coercer state cannot muster the political or military resources to communicate credible threats against the host state, triadic coercion is likely to fail. In most cases, it will not even be attempted. The schema’s second level concerns the host state’s predisposition toward the nonstate actor’s use of violence. All things being equal, “defiant” host states support nonstate attacks and “cooperative” hosts oppose them. States may be defiant or cooperative for a variety of reasons, including their own foreign policy goals, ideological orientation, domestic public opinion, or pressures to maintain a ruling coalition of key elites. We consider a predisposition in favor or against aiding nonstate actors to exist in advance of triadic coercion and regardless of its use. Practitioners and scholars devote considerable attention to host states’ predispositions. As we show in the Israeli case, decision-makers sometimes never move beyond the question of whether host states are defiant or cooperative. Similarly, key academic studies construct detailed typologies of host-states’ orientations toward nonstate violence. For example, Paul Pillar notes that a host state’s relationship to a terrorist group might be one of U N D E R S TA N D I N G T R I A D I C C O E R C I O N

[7]

,,17(567$7(%$/$1&(2)32:(5 &DQWKHFRHUFHUFUHGLEO\WKUHDWHQ WKHKRVW"